Supporting Content Contextualization in Web based Applications on
Mobile Devices
Alisa Sotsenko
1
, Marc Jansen
1,2
and Marcelo Milrad
1
1
Department of Media Technology, Linnaeus University, Vaxjo, Sweden
2
Institute of Computer Science, University of Applied Sciences Ruhr West, Mülheim, Germany
Keywords: Mobile Learning Applications, Context Detection, Cross Platform Development, Content Adaptation.
Abstract: Mobile devices, in the form of smartphones, are endowed with rich capabilities in terms of multimedia,
sensors and connectivity. The wide adoption of these devices allows using them across different settings and
situations. One area in which mobile devices become more and more prominent is within the field of mobile
learning. Here, mobile devices provide rich possibilities for the contextualization of the learner, by using the
set of sensors available in the device. On the one hand, the usage of mobile devices enables participation in
learning activities independent of time and space. Nevertheless, developing mobile learning applications for
the heterogeneity of mobile devices available in the market becomes a challenge. Not only this is a problem
related to form factor aspects, but also the large number of different operating systems, platforms and app
infrastructures (app stores) are aspects to be considered. In this paper we present our initial efforts with
regard to the development of cross-platform mobile applications to support the contextualization of learning
content.
1 INTRODUCTION
In recent years, the use of mobile devices to support
teaching and learning has gained increasing
attention. These devices have become important
components of modern learning environments and
they allow learners to get engaged in different
learning activities independent of time and space
(Sharples et al., 2009). Mobile devices, in the form
of smartphones or tablets, provide rich
functionalities with respect to the contextualization
of the learner as they incorporate a rich set of
sensors that provide data that can be used to gather
information about the current context of a user.
Determining the current geo-position of the learner
does provide a first step towards gathering the entire
set of contextual information of a learner. Other kind
of contextual information might include acceleration
sensor data that allows to track whether the learner is
currently on the move (e.g., in a public transport
system) or in a stable position (e.g., sitting in a café).
Furthermore, the digital compass available in
smartphones can be used to calculate the current
viewpoint of the learner. Additionally, the camera
can be used for capturing photos, and therefore
create additional learning materials or scan, e.g.,
barcodes providing access to certain learning
materials. Once the context is determined, the
different learning resources can be made available
and tailored to the learner´s situation through access
to standard learning management systems (LMS).
A major challenge on developing and scaling up
m-learning application addressing some of the issues
mention above is how to cope with the wide range of
mobile devices and operating systems available. We
agree with Thomas et al., (2012) in the sense that we
couldn’t follow an app-based strategy for core
services, at scale, with the uncertainty of all the time
having around a large number of different platforms
and changing markets. The development of mobile
applications aiming at providing a rich
contextualization for the learners is difficult because
it requires a large number of different platforms and
devices with different specifications (computational
power, screen-size, available memory, etc.) (Tan et
al., 2009). Kauninef et al., (2012) suggested a m-
learning approach that contextualize the learning
content by using multimedia capabilities of the
devices independently from the time and place.
Learning content becomes instead adaptable to the
device screen size and battery life. Other
approaches, such the one proposed by Giemza and
501
Sotsenko A., Jansen M. and Milrad M..
Supporting Content Contextualization in Web based Applications on Mobile Devices.
DOI: 10.5220/0004405005010504
In Proceedings of the 9th International Conference on Web Information Systems and Technologies (WEBIST-2013), pages 501-504
ISBN: 978-989-8565-54-9
Copyright
c
2013 SCITEPRESS (Science and Technology Publications, Lda.)
Jansen (2011) suggest the use of a flexible
architecture adaptable to the device screen, input
methods, etc. and provide different representation
views for the same content on different device types
in the learning scenarios.
One way to overcome some of the mentioned
above problems is the implementation of platform
independent mobile applications for retrieving
content and contextual information about the learner.
In this paper we describe our initial efforts while
exploring how to develop and implement different
mechanisms to support the contextualization of the
learner. The paper is organized as follows. Section
two introduces in details the problem description and
the motivation for this work. The last two sections
describe technical and implementations aspects and
the first implementation of a prototype. The paper
concludes with a short outlook and plans for future
work are presented.
2 PROBLEM DESCRIPTION
AND MOTIVATION
One of the challenges we are addressing is how
learning content in the form of learning objects,
materials, resources or services can be available and
retrieved in a way that they are relevant to the user
in his/her current context. Contextual information
about the environment (location, time, lightning,
noise, etc.) or communication resources (network
connectivity, communication costs, etc.) could be
used to determine some features of the user´s current
context (Anastasios, 2008). There are two main
problems associated to this situation; one of them is
how to provide the convenient access to the content
available at a LMS while the other one is taking into
account the wide range of mobile devices that
learners may have, so to provide the right features of
the user´s current context as discussed above.
According to the first problem, mobile devices
can be used to access learning material available at
different LMS (such as Moodle) via a mobile web
browser or by downloading native apps, which
usually do not take the current context of the learner
into account. Jordi et al., (2012) have recently
developed web service extensions to already existing
Moodle 2.0 web services that allow mobile
interactions with course content and management of
user's personal content, viewing and uploading
assignments, etc. Unfortunately, this extension does
not provide the possibility to download learning
materials (in formats such .pdf, .rtf, etc.) to the
mobile client. Therefore, to our knowledge there are
not many implementations able to provide the
suitable content depending on the user´s current
context. If it was possible to derive the user´s
contextual information combined with a set of rules
and behaviours, then we can offer and provide the
appropriate content and format of learning materials.
For example, students travelling on the train or bus
can use audio materials instead of text documents
due to poor lighting conditions. Contextualization of
the learning content in a variety of settings can make
the learning process more convenient.
One of the approaches to solve the second
problem related to software implementation issues is
to rely on web-based solutions. Using only
JavaScript, HTML and CSS provide us with limited
possibilities respect to taking advantage of the
device capabilities, mainly due to security
restrictions that do not allow access to device
specific hardware such as the camera, the
accelerometer, etc. In order to overcome this kind of
restrictions, solutions such as the PhoneGap
framework have been developed. With the help of
this framework, HTML5 and JavaScript applications
have the ability to use device specific resources on
mobile devices in a platform independent way. In
this paper we propose an approach that provides
convenient and flexible access to the content from
different LMS with the corresponding user´s current
context for different mobile devices. Since the
learners can have different types of mobile devices
there is a need for platform independent
development of such applications. In our case, this
purpose is achieved by the development of a hybrid
cross-platform mobile application that addresses the
issues discussed earlier in this section.
3 TECHNICAL APPROACH AND
IMPLEMENTATION ISSUES
This section describes the technical approach we
have chosen to implement the cross-platform mobile
application to support the contextualization of the
learner. In order to get the content (learning
materials) from a specific LMS (Moodle in our
case), we implemented a Web Service that directly
provides access to the learning resources (Fig. 1).
Therefore, single resources stored in the LMS can
easily be accessed by a learner by just scanning a
QR code. The idea behind using a QR code image is
for determining the current context (position, time
and so on) of the learner, as additional information
about the learning material.
WEBIST2013-9thInternationalConferenceonWebInformationSystemsandTechnologies
502
Figure 1: The different components of our proposed
approach.
On the server side, as shown in Fig.1, we have
implemented a Web Service that is capable of
transferring/interpreting images of scanned QR
codes taken by the camera of a mobile device.
Afterwards, the Web Service is capable of analysing
the images of the scanned QR code and it returns a
hash code. We decided not to directly store the URL
to the according resource in the QR code to increase
the flexibility of the provided solution. We have
implemented an additional Web Service that maps
the hash code to the relevant resource in the LMS.
By providing the necessary information gathered
by the sensors in the mobile device, the Web Service
can determine the current context in which the user
tries to access a certain learning material. For
example, with the help of the geo-position of the
user and the values measured by the acceleration
sensor, it could be possible to determine where the
learner is currently located and if he/she is moving
or not. Furthermore, the selection of learning
materials might be influenced by the contextual
information about the learner. If the learner is
currently moving or located on a nearly fixed
position, e.g., a podcast of the corresponding
learning material might be more appropriate than a
presentation on slides since we might assume that
the learner is currently driving in his car or on a
public transportation system. The other way round,
if the learner is in a geographically fixed position, it
might be assumed that he/she is currently sitting in a
particular location where consuming visual materials
(slides or video presentation) might be appropriate.
Therefore, learning materials can be selected (based
on a set of recommendations) for supporting content
contextualization. For instance, for each set of
possible defined scenarios we could provide a set of
recommendations about learning material (e.g.
format type) with respect to current context/position.
As described above, all major functionalities are
implemented as Web Services that allows rapid and
flexible development of new applications that make
use of the provided functionality.
There are several approaches for developing
cross-platform mobile application. One of them is to
use specific development frameworks that provide a
platform-independent API where the code, using a
cross-compiler, is transformed into platform-specific
native app targeted at the different platforms that the
application will run on. The advantages of this
approach are the access of platform dependent
resources of the mobile device and an increased
performance, as the application works natively on
the device. On the other hand, the disadvantage of
this approach is the complexity of writing the cross
compiler. There is another approach to develop
mobile web applications that will run in a browser-
view embedded into a native app as hybrid apps.
PhoneGap is a framework that utilizes this approach
and provides a cloud service that easy can compile
and build mobile apps with the latest version of the
SDK for the target platforms. One of the main
advantages of this approach is related to
performance, as it runs on the mobile device not in
the browser and processes the JavaScript locally.
4 IMPLEMENTATION OF THE
FIRST PROTOTYPE
This section describes the implementation of the first
prototype and the functions and interactions between
the Web Services and the mobile application.
Figure 2: The operation of retrieving information about the
resources from the LMS.
A learner can take a picture of the QR code
image as described in the previous section. After this
the Web Service that translates the hash code into
information about the resources of the LMS is
called. For instance, information about different
format types of learning material, path to those in
the LMS, as well as name, size, etc., could be
retrieved from this service. After getting additional
information about the learning material, this content
can be retrieved from the LMS by using the Web
Service implemented for providing the suitable
SupportingContentContextualizationinWebbasedApplicationsonMobileDevices
503
resources from the LMS (see Figure 2 above).
For the implementation of the client side, we
used the Vaadin TouchKit framework that allows
creating the mobile user interface similar to native
mobile apps. This framework gives the opportunity
to create the mobile application on Java without
using any scripting languages. This framework
doesn’t provide the creation of hybrid applications
for that we used the PhoneGap Build service for
building the native app for the target platforms.
To
get access to the specific hardware of the mobile
device (e.g., the camera) we used the PhoneGap
library, made for using it in Google Web Toolkit
(GWT) applications) that provides access to almost
all mobile features. Furthermore, new features are
flexibly added to this API so that the integration is
possible seamlessly.
5 CONCLUSIONS
This paper described an approach for receiving
contextualized learning materials from a learning
management system on mobile devices with
gathering additional information about the learner’s
current context. One of the major benefits of the
presented approach is the platform independent
implementation of the mobile application by using
PhoneGap build service that allows customized user
interfaces best suited for the different platforms we
wanted to support. Furthermore, the proposed
approach provides a great deal of flexibility by
implementing the major features as stand-alone Web
Services that can easily be consumed also by other
applications.
Future plans for the development of our work do
include on the one hand, a heuristic evaluation
focussing on the model and software engineering
approach used for retrieving the contextual
information and defining the set of
recommendations according to this information.
Particularly identify the main context dimensions,
design and explore multi-dimensional vector-space
model for contextualization of learner. This will
include using not only the camera on the mobile
device but also additional sensors as GPS,
accelerometer, etc. On the other hand, a comparison
evaluating the advantages and drawbacks of the
different frameworks used for the implementation of
cross platform mobile apps, as presented earlier is
also planned. We are currently deploying the mobile
application using IBM´s mobile development
platform called IBM worklight studio.
REFERENCES
Anastasios, A., (2008). Context-Aware Mobile Learning.
Greece, September. In proceedings of the Conference
on Communications in Computer and Information
Science (CCIS) 19, 213-220.
Jordi, P., Marc, A., Maria, J.C., Enric, M., Nikolas, G.,
(2012). Moodbile: a Moodle web service extension for
mobile applications. In Proceedings of First Moodle
Research Conference, 146-155
Giemza, A. & Jansen, M. (2011). An Architectural
Approach to Support Multi-Device Learning
Environments. In: Proceedings of the IADIS
Conference on Mobile Learning 2011, Avila, Spain
Kauninef, B., Tlemsani, R., Rerbal, S.M. Lotfi, A. (2012).
Developing a Mobile Learning Approach in Platform
LMS INT TIC. Information Technology Journal,
11(8), 1133-1137
Sharples, M., Amedillo Sanchez, I., Milrad, M., &
Vavoula , G. (2009). Mobile learning: small devices,
big issues. In: Balacheff, N.; Ludvigsen, S.; Jong , T.
de and Barnes , S. eds. Technology Enhanced
Learning: Principles and Products. Heidelberg,
Germany: Springer, 233–249.
Thomas, R. (2012). OU Mobile VLE: extending the reach
of studying through the mobile web. Proceeding of
Moodle Research Conference, Heraklion, Crete, 14-
15.
Tan, Q., Kinshuk , Yen-Hung, K., Yu-Lin, J.,Po-Han,
W.,Yueh-Min,H.,Tzu-Chien, L., & Maiga, C. (2009).
Location-Based Adaptive Mobile Learning Research
Framework and Topics. In proceedings of the
Conference on Computational Science and
Engineering, 140-147
WEBIST2013-9thInternationalConferenceonWebInformationSystemsandTechnologies
504